Cedar and cypress pollen counts are associated with the prevalence of allergic diseases in Japanese schoolchildren

Authors


  • Edited by: Wytske Fokkens

Correspondence

Koichi Yoshida, Tokyo Metropolitan Children's Medical Center, 2-8-29 Musashidai, Fuchu, Tokyo, Japan.

Tel.: +81-42-300-5111

Fax: +81-42-312-8163

E-mail: kouichi_yoshida@tmhp.jp

Abstract

Background

Patients allergic to pollen have been known to become more symptomatic during pollen season compared with the nonpollen season. However, there are few studies regarding whether higher exposure to pollen might increase the prevalence of allergic diseases.

Methods

An ecological analysis was conducted to evaluate whether pollen exposure is associated with the prevalence of allergic diseases in schoolchildren. Pollen count data of Japanese cedar (Cryptomeria japonica) and Japanese cypress (Chamaecyparis obtusa), which are the major pollen allergens in Japan, were obtained from each prefecture. The prevalence of allergic diseases in schoolchildren in each prefecture was based on a nationwide cross-sectional survey using the International Study of Asthma and Allergies in Childhood questionnaire.

Results

After omitting three prefectures where pollen data were not available, data of 44 prefectures were analysed. The prevalence of allergic rhinoconjunctivitis in children aged 6–7 years was positively associated with both cedar and cypress pollen counts (P = 0.01, both), whereas the prevalence of allergic rhinoconjunctivitis in children aged 13–14 years was positively associated with only cypress pollen counts (P = 0.003). Furthermore, the prevalence of asthma was positively associated with cedar pollen counts in 6- to 7-year-old children (P = 0.003) but not cypress pollen counts in either age group.

Conclusions

There are ecological associations between pollen counts and the prevalence of allergic diseases in Japanese schoolchildren. Further studies are needed to determine whether the difference between the effects of cedar and cypress pollens is attributable to pollen counts or allergenicity.

Asthma and allergic rhinitis are the most common chronic diseases in childhood and impair the quality of life of the patients and their family [1, 2]. The incidence of both disorders has increased dramatically worldwide in the last few decades [3]. However, the prevalence of allergic diseases varies widely throughout the world. The International Study of Asthma and Allergies in Childhood (ISAAC) showed that the prevalence of asthma and allergic rhinoconjunctivitis varies 20- to 40-fold in the world [4, 5].

Pollens, along with climate and air pollutants, are among the environmental factors hypothesized to contribute to this variation [6]. In patients with allergic rhinitis, symptoms become more frequent and more severe when pollen counts increase [7, 8]. Furthermore, patients with asthma require attendance at emergency departments and hospital admissions more frequently when the airborne pollen concentration is higher [9-12].

However, there are only a few studies evaluating the ecological associations between pollen counts and the prevalence of allergic diseases, and the results were inconsistent. Studies comparing the prevalence of allergic diseases between an area with higher pollen count and another with low pollen count showed that higher exposure to pollen was associated with a higher sensitization rate in children [13] and prevalence of allergic rhinitis in adults [14, 15]. By contrast, a large study performed in 28 centres within 11 countries showed that there was little relationship between pollen exposure and the prevalence of allergic symptoms in children [16]. This inconsistency may be attributable to geographical differences in the major pollen species and lifestyles of the study subjects. Ecological studies to evaluate this association for specific pollen exposure in a large homogeneous population are warranted.

Therefore, we performed an ecological analysis to evaluate whether there were associations between specific pollen counts (Japanese cedar and Japanese cypress) and the prevalence of allergic diseases in Japanese schoolchildren, using data on pollen counts and the prevalence of allergic diseases in each prefecture throughout Japan.

Methods

Study participants

The prevalence of allergic diseases in children aged 6–7 and 13–14 years in each prefecture was based on the data of a nationwide survey that was conducted throughout Japan in 2008; details of the methods and response rates have already been published [17]. In this survey, samples were randomly selected from all prefectures (n = 47) in Japan using public schools as the sampling units because more than 95% of schoolchildren attend public schools in Japan.

Questionnaire

The survey used the Japanese version of the written questionnaire of ISAAC, which was distributed by teachers at the participating schools [17]. The responses to the questions were reported by parents for children aged 6–7 years and were self-reported for children aged 13–14 years.

Allergic rhinoconjunctivitis was defined as positive answers to both of these questions: ‘In the past 12 months, have you (has your child) had a problem with sneezing, or a runny, or blocked nose when you (he/she) did not have a cold or the flu?’ and ‘In the past 12 months, has this nose problem been accompanied by itchy-watery eyes?’ Asthma was defined as a positive answer to the question: ‘Have you (has your child) had wheezing or whistling in the chest in the past 12 months?’

Pollen and meteorological data

Japanese cedar (Cryptomeria japonica, family Taxodiaceae) and Japanese cypress (Chamaecyparis obtusa, family Cupressaceae) are the major causes of pollinosis in Japan. We used pollen data from the Association of Pollen Information in Japan. The cedar and cypress pollen counts were measured at observation facilities located in all prefectures except for Okinawa. Because a Durham's sampler has been the most popular apparatus for measuring pollen counts in Japan, we omitted data from two prefectures in which the pollen counts were measured by different methods. The data from these three prefectures were excluded, and the data from the 44 remaining prefectures included in the final analysis. The average values of the pollen counts per year over the past 4 years were used in this study because it may take time for children to become sensitized and develop allergic symptoms [8, 18]. Meteorological data were obtained from the Japan Meteorological Agency (http://www.jma.go.jp/jma/indexe.html). The mean annual temperature and relative humidity measured at each prefectural capital in 2008 were used in this analysis.

Statistical analyses

Chi-square tests were used to assess whether the prevalence of allergic diseases differed between children included in this study and those who were excluded. The data for the two types of tree pollen and the two age groups were analysed separately. Associations between pollen counts and the prevalence of allergic rhinoconjunctivitis were determined using Pearson's product-moment correlation coefficient. Multivariable regression analyses of the associations between pollen counts and the prevalence of allergic symptoms were adjusted for the gender ratio [17], mean annual temperature [19] and mean annual relative humidity [19]. The associations were analysed after adjustment for the pollen counts of other species as potential confounders, as cross-reactivity exists between cedar and cypress pollens [20]. The prevalence of allergic rhinoconjunctivitis was included as a confounder in the analyses of the associations between pollen counts and the prevalence of asthma because allergic rhinitis may affect asthma patients [21, 22]. P values <0.05 were considered to indicate statistical significance. All analyses were performed using the statistical package ‘SPSS for Windows version 19’ (SPSS Inc., Chicago, IL, USA).

Ethics

The study protocol was approved by the independent review board of the National Center for Child Health and Development.

Results

Data were analysed in 44 prefectures, in which cedar and cypress pollen counts were measured separately, including 40 975 children aged 6–7 years and 45 787 children aged 13–14 years. The average values of cedar and cypress pollen counts in the 44 prefectures analysed were 2967 counts/cm2 (range, 34–7912 counts/cm2) and 1245 counts/cm2 (range, 1–6048 counts/cm2), respectively. Cedar pollen counts were higher in prefectures in eastern Japan (Fig. 1A), and cypress pollen counts were higher in prefectures on the Pacific side than along the Sea of Japan (Fig. 1B).

Figure 1.

Maps of pollen counts and the prevalence of allergic rhinoconjunctivitis in Japan. (A) Japanese cedar pollen counts. (B) Japanese cypress pollen counts. (C) The prevalence of allergic rhinoconjunctivitis in 6- to 7-year-old children. (D) The prevalence of allergic rhinoconjunctivitis in 13- to 14-year-old children.

There was a wide range of the prevalence of allergic rhinoconjunctivitis (range, 8.1–29.2%) and asthma (range, 9.4–17.3%) between prefectures in the 6- to 7-year-old children. Similar to the 6- to 7-year-old children, the prevalence of allergic rhinoconjunctivitis ranged from 10.8% to 30.9% and that of asthma ranged from 6.1% to 13.2% in the 13- to 14-year-old children. The prevalence of allergic rhinoconjunctivitis in 6- to 7-year-old children was higher in prefectures in eastern Japan (Fig. 1C), whereas the prevalence of allergic rhinoconjunctivitis in 13- to 14-year-old children was higher in prefectures on the Pacific side of eastern and central Japan (Fig. 1D). Cedar pollen counts were positively associated with the prevalence of allergic rhinoconjunctivitis in 6- to 7-year-old children (R = 0.48, P = 0.001) (Fig. 2A) but not in 13- to 14-year-old children (R = 0.18, P = 0.24) (Fig. 2B). Cypress pollen counts were positively associated with the prevalence of allergic rhinoconjunctivitis in 6- to 7-year-old children (R = 0.43, P = 0.004) and 13- to 14-year-old children (R = 0.47, P = 0.001) (Fig. 2C,D). Even after adjustment for confounders, the prevalence of allergic rhinoconjunctivitis remained positively associated with cedar pollen counts for the 6- to 7-year-old children (P = 0.01) and cypress pollen counts for both the 6- to 7-year-old children and 13- to 14-year-old children (P = 0.01 and P = 0.003, respectively) (Table 1). Cedar pollen counts were not associated with the prevalence of allergic rhinoconjunctivitis in 13- to 14-year-old children after adjustment for confounders (P = 0.29).

Table 1. Associations between pollen counts and the prevalence of allergic diseases
 Japanese cedarJapanese cypress
Coefficient (SE)P-valueCoefficient (SE)P-value
  1. SE, standard error.

  2. Coefficient is for each pollen count increment of 1000 counts/cm2.

  3. a

    Adjusted for the gender ratio, mean annual temperature, mean annual relative humidity, and cypress pollen counts.

  4. b

    Adjusted for the gender ratio, mean annual temperature, mean annual relative humidity, and cedar pollen counts.

  5. c

    Adjusted for the gender ratio, mean annual temperature, mean annual relative humidity, cypress pollen counts, and prevalence of allergic rhinoconjunctivitis.

  6. d

    Adjusted for the gender ratio, mean annual temperature, mean annual relative humidity, cedar pollen counts, and prevalence of allergic rhinoconjunctivitis.

Allergic rhinoconjunctivitis
6- to 7-year-old children1.07 (0.39)a0.011.49 (0.57)b0.01
13- to 14-year-old children0.34 (0.32)a0.291.52 (0.49)b0.003
Asthma
6- to 7-year-old children0.49 (0.16)c0.003−0.43 (0.23)d0.07
13- to 14-year-old children0.11 (0.15)c0.460.04 (0.30)d0.89
Figure 2.

Pollen counts and the prevalence of allergic rhinoconjunctivitis (ARC). (A) Positive association between cedar pollen counts and the prevalence of ARC in 6- to 7-year-old children (R = 0.48, P = 0.001). (B) No association between cedar pollen counts and the prevalence of ARC in 13- to 14-year-old children (R = 0.18, P = 0.24). (C) Positive association between cypress pollen counts and the prevalence of ARC in 6- to 7-year-old children (R = 0.43, P = 0.004). (D) Positive association between cypress pollen counts and the prevalence of ARC in 13- to 14-year-old children (R = 0.47, P = 0.001).

In general, the prevalence of allergic rhinoconjunctivitis was higher in 13- to 14-year-old children than in 6- to 7-year-old children. However, the difference between the two age groups was inversely associated with the prevalence of allergic rhinoconjunctivitis in the younger children (R = −0.52, P < 0.001) (Fig. 3A). Therefore, we analysed the association between the pollen counts and the differences in the prevalence of allergic rhinoconjunctivitis between the two age groups. Differences in the prevalence of allergic rhinoconjunctivitis were inversely associated with cedar pollen counts (R = −0.50, P = 0.001) (Fig. 3B) but not with cypress pollen counts (R = −0.28, P = 0.86) (Fig. 3C).

Figure 3.

Differences in the prevalence of allergic rhinoconjunctivitis (ARC) between 6- to 7-year-old children and 13- to 14-year-old children. (A) Inverse association between the prevalence of ARC in 6- to 7-year-old children and the difference in the prevalence of ARC between the two age groups (R = −0.52, P < 0.001). (B) Inverse association between cedar pollen counts and the difference in the prevalence of ARC between the two age groups (R = −0.50, P = 0.001). (C) No association between cypress pollen counts and the difference in the prevalence of ARC between the two age groups (R = −0.03, P = 0.86).

We next analysed the associations between pollen counts and the prevalence of asthma (Table 1). After adjustment for confounders, cedar pollen counts were positively associated with the prevalence of asthma in 6- to 7-year-old children (P = 0.003) but not in 13- to 14-year-old children (P = 0.46). Cypress pollen counts were not associated with the prevalence of asthma in either age group (P = 0.07 for the 6- to 7-year-old children and P = 0.89 for the 13- to 14-year-old children).

Discussion

In this ecological study, we found a positive association between cedar and cypress pollen counts and the prevalence of allergic rhinoconjunctivitis and asthma in Japanese schoolchildren. Consistent with our finding, a study performed in Italian children aged 11–14 years revealed that children living in a high-pollen-count area showed a significantly higher percentage of sensitization to pollens than those in another area with low pollen counts [13]. Similar results were shown in French adults [15] and the genetically homogeneous Inuit population [14], although their sample sizes were small. By contrast, an ecological study performed in 11 countries (9 European countries, Australia and Kuwait) revealed that there was little relationship between pollen exposure and the prevalence of allergic symptoms in children aged 13–14 years [16]. Inconsistency with our results might be explained by the geographical heterogeneity in the lifestyle of the study subjects [23, 24] and the prevalence of plant species and their related allergens [25].

Cedar pollen counts were positively associated with the prevalence of allergic rhinoconjunctivitis in 6- to 7-year-old children but not in 13- to 14-year-old children. Although the prevalence of allergic rhinoconjunctivitis is generally higher in 13- to 14-year-old children than in 6- to 7-year-old children [5], the differences in the prevalence of allergic rhinoconjunctivitis between the two age groups were inversely associated with the cedar pollen counts. A retrospective analysis performed in the United States revealed that over 50% of children with allergic rhinitis were sensitized to at least one pollen by the age of 3 years and that the sensitization rate increased with age and plateaued by the age of 8 years [18]. Together with our results, it is suggested that children in areas heavily exposed to cedar pollen might be sensitized to cedar pollen in early childhood, and the prevalence of allergic rhinoconjunctivitis might therefore plateau by the age of 6–7 years. By contrast, the prevalence of allergic rhinoconjunctivitis in less-exposed areas might not yet have plateaued by the age of 6–7 years and might thus continue to increase thereafter. Consequently, cedar pollen counts were not associated with the prevalence of allergic rhinoconjunctivitis in 13- to 14-year-old children.

Unlike cedar pollen counts, cypress pollen counts were positively associated with the prevalence of allergic rhinoconjunctivitis in 6- to 7-year-old children, and this positive association persisted in 13- to 14-year-old children. In Japan, cedar pollen counts are usually more than twice those of cypress. Therefore, the prevalence of allergic rhinoconjunctivitis due to cypress pollen might require additional time to reach a plateau than that caused by cedar pollen. The discrepancy between the results for cedar and cypress pollens may be attributable not only to differences in pollen counts but also to differences in antigenicity. T-cell reactivity differs between cedar and cypress pollens [26, 27], although there is some cross-reactivity [20].

Cedar pollen counts were positively associated with the prevalence of asthma in 6- to 7-year-old children even after adjustment for confounders, including the prevalence of allergic rhinoconjunctivitis. Pollens may induce asthma symptoms independently of allergic rhinitis by two mechanisms. The first is inhalation of pollen allergens. Pollen grains are generally too large to penetrate the lower airway and thus do not provoke asthma symptoms. However, small particles of cedar pollen contain a major pollen allergen (Cry j 1) [28] and are likely to induce an asthma attack [29]. The second is that pollens may act as adjuvants to exacerbate asthmatic symptoms. Intranasal administration of ovalbumin with cedar pollen induced ovalbumin-specific IgE responses, although the administration of ovalbumin alone did not induce the production of ovalbumin-specific IgE [30]. Cedar pollen may thus enhance sensitization to other allergens as well as pollen itself and thereby influence asthma in young children. The prevalence of asthma was not associated with cypress pollen counts. The reason for this discrepancy between cedar and cypress pollens remains unclear and warrants further investigation.

The strength of our study is that it addresses the associations between two types of tree pollen and the prevalence of allergic diseases in children in two different age groups. One limitation is that our study was a questionnaire-based survey without testing for sensitization. Estimation of the prevalence of allergic rhinoconjunctivitis by a questionnaire only may be not very sensitive in young children [31]. However, sensitization to any allergen is strongly associated with allergic rhinoconjunctivitis as assessed by the ISAAC questionnaire [32], and this questionnaire has previously been used for ecological analyses [6, 16]. Another limitation is that we did not adjust our analysis for the levels of air pollutants, such as SPM, SO2 and NOX. Air pollutants can affect both allergic subjects [33] and the allergenicity of pollens [34]. The levels of these pollutants have been reported to be affected by the distance from major roads and the traffic count [35] and vary widely even within the same prefecture. Therefore, we did not include air pollutants as confounders in this analysis.

In conclusion, pollen counts of cedar and cypress are positively associated with the prevalence of allergic rhinoconjunctivitis and asthma in Japanese schoolchildren. Although both cedar and cypress pollens are tree pollens, they show different effects regarding the prevalence of allergic diseases. Further studies are required to elucidate the reason for this discrepancy.

Acknowledgments

We would like to thank all of the students and parents who participated in this survey. We are indebted to Mari Sasaki for critical reading of the manuscript.

Funding

This study was supported by a Health and Labour Sciences Research Grant for Research on Allergic Disease and Immunology from the Ministry of Health, Labour, and Welfare, Japan.

Author contributions

Koichi Yoshida analysed the data and wrote the article. Yuichi Adachi and Akira Akasawa designed the study protocol and cowrote the article. Masayuki Akashi, Yukihiro Ohya and Hiroshi Odajima designed the study protocol. Toshiko Itazawa and Yoko Murakami analysed the data and discussed the results.

Conflict of interest

All authors declare that there are no conflicts of interest.

Ancillary